BDC Publication List (2001 - )
[[ Book Chapter ]]
A connectionist model which unifies the behavioral and the linguistic processes: Results from robot learning experiments (2002)
Yuuya Sugita and Jun Tani
Mirror Neurons and the Evolution of Brain and Language (Edited by M.I.Stamenov and V.Gallese, Advances in Consciousness Research Vol.42, pp. 363-376)
[ last update: 2003/11/22 ]
Symbols and Dynamics in Embodied Cognition: Revisiting a Robot Experiment (2003)
Jun Tani
In (Eds) Butz M. V., Sigaud O., Gerard P., Anticipatory Behavior in Adaptive Learning Systems, Springer-Verlag, pp. 167-178
[ last update: 2003/12/10 ]
[[ Journal ]]
On the dynamics of robot exploration learning (2002)
J. Tani and J. Yamamoto
Cognitive Systems Research, Vol.3, No.3, pp.459-470
[ last update: 2005/02/14 ]
Learning to generate articulated behavior through the bottom-up and the top-down interaction processes (2003)
Jun Tani
Neural Networks, Vol.16, No.1, pp. 11-23
[ last update: 2003/12/10 ]
Self-Organization of Behavioral Primitives as Multiple Attractor Dynamics: A Robot Experiment (2003)
Jun Tani and Masato Ito
IEEE Transactions on Systems, Man, and Cybernetics Part A: Systems and Humans, Vol.33, No.4, pp. 481-488
[ last update: 2003/12/10 ]
Learning Semantic Combinatoriality from the Interaction between Linguistic and Behavioral Processes (2004)
Yuuya Sugita and Jun Tani
Adaptive Behavior (in press)
- Abstract
We present a novel connectionist model for acquiring the semantics of a simple language through the behavioral experiences of a real robot. We focus on the ``compositionality'' of semantics and examine how it can be generated through experiments. Our experimental results showed that the essential structures for situated semantics can self-organize themselves through dense interactions between linguistic and behavioral processes whereby a certain generalization in learning is achieved. Our analysis of the acquired dynamical structures indicates that an equivalence of compositionality appears in the combinatorial mechanics self-organized in the neuronal nonlinear dynamics. The manner in which this mechanism of compositionality, based on dynamical systems, differs from that considered in conventional linguistics and other synthetic computational models, is discussed in this paper. \abstract:
- Downloadable Materials
[ last update: 2004/09/10 ]
Learning to generate combinatorial action sequences utilizing the initial sensitivity of deterministic dynamical systems (2004)
R. Nishimoto and J. Tani
Neural Networks, Vol.17, pp.925-933
[ last update: 2005/02/14 ]
Motor primitive and sequence self-organization in a hierarchical recurrent neural network (2004)
R. W. Paine and J. Tani
Neural Networks, Vol.17, pp.1291-1309
[ last update: 2005/02/14 ]
On-line Imitative Interaction with a Humanoid Robot Using a Dynamic Neural Network Model of a Mirror System (2004)
M. Ito and J. Tani
Adaptive Behavior, Vol.12, No.2, pp.93-115
[ last update: 2005/02/14 ]
Open-end human robot interaction from the dynamical systems perspective: Mutual adaptation and incremental learning (2004)
T. Ogata, S. Sugano, and J. Tani
Lecture Notes in Artificial Intelligence, Vol.3029, pp.435-444
[ last update: 2005/02/14 ]
Self-organization of distributedly represented multiple behavior schemata in a mirror system: reviews of robot experiments using RNNPB (2004)
J. Tani, M. Ito, and Y. Sugita
Neural Networks, Vol.17, pp.1273-1289
[ last update: 2005/02/14 ]
The dynamical systems accounts for phenomenology of immanent time: An interpretation by revisiting a robotics synthetic study (2004)
Jun Tani
Journal of Consciousness Studies, Vol.11, No.9, pp.5-24
[ last update: 2005/02/14 ]
Undecidability in the Imitation Game (2004)
Yuzuru Sato and Takashi Ikegami
Minds and Machines, Vol.14, pp. 133-143
[ last update: 2004/04/21 ]
Acquisition of Motion Primitives of Robot in Human-Navigation Task (2005)
T. Ogata, S. Sugano, and J. Tani
Journal of Transactions of the Japanese Society for Artificial Intelligence, Vol.20, No.3, pp.188-196
[ last update: 2005/04/14 ]
Extracting Multimodal Dynamics of Objects using RNNPB (2005)
T. Ogata, H. Ohba, J. Tani, K. Komatani, and H. G. Okuno
Journal of Robotics and Mechatronics, Vol.17, No.6, pp.681-688
[ last update: 2006/06/14 ]
How Hierarchical Control Self-organizes in Artificial Adaptive Systems (2005)
R. W. Paine and J. Tani
Adaptive Behavior, Vol.13, No.3, pp.211-225
[ last update: 2006/06/14 ]
Open-end human-robot interaction from the dynamical systems perspective: mutual adaptation and incremental learning (2005)
T. Ogata, S. Sugano, and J. Tani
Advanced Robotics, Vol.19, No.6, pp.651-670
[ last update: 2006/06/14 ]
Dynamic and interactive generation of object handling behaviors by a small humanoid robot using a dynamic neural network model (2006)
M. Ito, K. Noda, Y. Hoshino, and J. Tani
Neural Networks, Vol.19, pp.323-337
[ last update: 2007/06/06 ]
[[ Conference Proceedings ]]
Articulation of Sensory-Motor Experiences by "Forwarding Forward Model" From Robot Experiments to Phenomenology (2002)
Jun Tani
Proc. of the 7th Int. Conf. on Simulation of Adaptive Behavior (SAB'02), MIT Press, pp. 171-180
[ last update: 2004/01/15 ]
Self-organization of behavioral primitives as multiple attractor dynamics: a robot experiment (2002)
J. Tani
Proc. 2002 Int. Joint. Conf. on Neural Networks (IJCNN'02), pp.489-494
[ last update: 2005/06/21 ]
The level organizatio by `Forwarding Forward models': from robot experiments (2002)
J. Tani
Proc. 7th Int. Symp. on Artificial Life and Robotics (AROB 7th'02), pp.359-366
[ last update: 2005/06/21 ]
A Holistic Approach to Compositional Semantics: a connectionist model and robot experiments (2003)
Yuuya Sugita and Jun Tani
Advances in Neural Information Processing Systems 2003 (NIPS*2003)
- Abstract
We present a novel connectionist model for acquiring the semantics of a simple language through the behavioral experiences of a real robot. We focus on the ``compositionality'' of semantics, a fundamental characteristic of human language, which is the ability to understand the meaning of a sentence as a combination of the meanings of words. We also pay much attention to the ``embodiment'' of a robot, which means that the robot should acquire semantics which matches its body, or sensory-motor system. The essential claim is that an embodied compositional semantic representation can be self-organized from generalized correspondences between sentences and behavioral patterns. This claim is examined and confirmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors.
- Downloadable Materials
[ last update: 2004/01/29 ]
Collaboration Development through Interactive Learning between Human and Robot (2003)
Tetsuya Ogata, Noritaka Masago, Shigeki Sugano, and Jun Tani
3rd International Workshop on Epigenetic Robotics, pp. 99-106
[ last update: 2003/12/22 ]
Interactive Learning in Human-Robot Collaboration (2003)
Tetsuya Ogata, Noritaka Masago, Shigeki Sugano, and Jun Tani
IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS'2003), pp.162-167
[ last update: 2003/12/22 ]
A Connectionist Approach to Learn Association between Sentences and Beha vioral Patterns of a Robot (2004)
Yuuya Sugita and Jun Tani
Proc. of the 8th International Conference on Simulation of Adaptive Behavior (SAB'04)
- Abstract
We focus on the ``compositionality'' of semantics, a fundamental characteristic of human language, which is the ability to understand the meaning of a sentence as a combination of the meanings of words. We also pay much attention to the ``embodiment'' of a robot, which means that the robot should acquire semantics which matches its body, or sensory-motor system. The essential claim is that an embodied compositional semantic representation can be self-organized from generalized correspondences between sentences and behavioral patterns. This claim is examined and confirmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors.
- Downloadable Materials
[ last update: 2004/09/10 ]
Acquisition of Motion Primitives of Robot in Human-Navigation Task: Towards Human-Robot Interaction based on "Quasi-Symbol" (2004)
T. Ogata, S. Sugano, and J. Tani
Proc. 2nd Int. Workshop on Man-Machine Symbiotic Systems, pp.315-326
[ last update: 2005/06/21 ]
Adaptive Motor Primitive and Sequence Formation in a Hierarchical Recurrent Neural Network (2004)
Rainer W. Paine and Jun Tani
Schaal, S., et al. (Eds.): From Animals to Animats 8: Proceedings of the Eigth International Conference on the Simulation of Adaptive Behavior (SAB 2004). MIT Press, Bradford Books, Cambridge, MA, pp. 274-283.
- Abstract
This study describes how complex goal-directed behavior can be obtained through adaptation in a hierarchically organized recurrent neural network using a genetic algorithm. Robot simulations showed that different types of dynamic structures self-organize in the lower and higher levels of the network for the purpose of achieving complex navigation tasks. Behavior primitives are switched in a top-down way through lower level parametric bifurcation structures. In the higher level, a topologically ordered mapping of initial cell activation states to motor-primitive sequences self-organizes by utilizing the initial sensitivity characteristics of nonlinear dynamical systems. The biological plausibility of the model’s essential principles is discussed.
- Downloadable Materials
[ last update: 2005/03/18 ]
Evolved Motor Primitives and Sequences in a Hierarchical Recurrent Neural Network (2004)
Rainer W. Paine and Jun Tani
Deb, K., et al. (Eds.): Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2004). Lecture Notes in Computer Science 3102. Springer-verlag, Berlin, pp. 603-614
- Abstract
This study describes how complex goal-directed behavior can evolve in a hierarchically organized recurrent neural network controlling a simulated Khepera robot. Different types of dynamic structures self-organize in the lower and higher levels of a network for the purpose of achieving complex navigation tasks. The parametric bifurcation structures that appear in the lower level ex-plain the mechanism of how behavior primitives are switched in a top-down way. In the higher level, a topologically ordered mapping of initial cell activa-tion states to motor-primitive sequences self-organizes by utilizing the initial sensitivity characteristics of nonlinear dynamical systems. A further experi-ment tests the evolved controller’s adaptability to changes in its environment. The biological plausibility of the model’s essential principles is discussed.
- Downloadable Materials
[ last update: 2005/03/18 ]
Generalization in Learning Multiple Temporal Patterns Using RNNPB (2004)
M. Ito and J. Tani
N. R. Pal et al. (Eds.): Proc. 11th Int. Conf. on Neural Information Processing (ICONIP2004), Lecture Notes in Comuter Science 3316, pp.592-598, Springer-Verlag
[ last update: 2005/06/21 ]
Human-Robot Collaboration Using Behavioral Primitives (2004)
T. Ogata, M. Matsunaga, S. Sugano, and J. Tani
Proc. 2004 IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS2004), pp.1592-1597
[ last update: 2005/06/21 ]
Interacting with NeuroCognitive Robots: A Dynamical Systems View (2004)
J. Tani and M. Ito
Proc. 2nd Int. Workshop on Man-Machine Symbiotic Systems, pp.123-134
[ last update: 2005/06/21 ]
Joint attention between a humanoid robot and users in imitation game (2004)
M. Ito and J. Tani
Proc. 3rd Int. Conf. on Development and Learning (ICDL'04)
[ last update: 2005/06/21 ]
On-line imitative interaction with a humanoid robot using a mirror neuron model (2004)
M. Ito and J. Tani
Proc. 2004 IEEE Int. Conf. on Robotics & Automation (ICRA2004), pp.1071-1076
[ last update: 2005/06/21 ]
Review of a dynamic neural network scheme for synthesizing cognition of robots and humanoids (2004)
J. Tani, M. Ito, and Y. Sugita
Proc. IEEE-RAS/RSJ Int. Conf. on Humanoid Robots (Humanoids 2004)
[ last update: 2005/06/21 ]
[[ Conference Talk ]]
[[ BDC Technical Report ]]
Learning to Generate Articulated Behavior Through the Bottom-Up and the Top-Down Interaction Processes (BDC-TR-2001-1)
Jun Tani
[ last update: 2003/11/11 ]
On the Dynamics of Robot Exploration Learning (BDC-TR-2001-2)
Jun Tani
[ last update: 2003/11/11 ]
Computation with Switching map systems: Nonlinearity and computational complexity (BDC-TR-2001-3)
Sato, Y., Taiji, M. and Ikegami, T
[ last update: 2003/11/11 ]
Undecidability in the imitation game (BDC-TR-2001-4)
Sato, Y. and Ikegami, T.
[ last update: 2003/11/11 ]
Chaos in learning a simple two person game (BDC-TR-2001-5)
Sato, Y., Akiyama, E. and Farmer, J. D.
[ last update: 2003/11/11 ]
Dynamics of learning and coupled replicator equations (BDC-TR-2001-6)
Sato, Y. and Crutchfield, J. P.
[ last update: 2003/11/11 ]
Collaboration Development through intaractive Learning between Human and Robot (BDC-TR-2003-1)
Tetsuya OGATA, Noritaka MASAGO, Shigeki SUGANO, and Jun TANI
[ last update: 2003/11/11 ]
Recognizing Sentences from Behavioral Experiences of a Robot (BDC-TR-2003-2)
Yuuya Sugita, and Jun Tani
- Abstract
We present a novel connectionist model for acquiring the semantics of language through the behavioral experiences of a real robot. We focus on the ``compositionality'' of semantics, which is a fundamental characteristic of human language that we can understand the meaning of a sentence as a combination of the meanings of words. The essential claim is that a compositional semantic representation can be self-organized from generalized correspondences between sentences and behavioral patterns. This claim is examined and confirmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors.
- Downloadable Materials
[ last update: 2003/11/11 ]
Generalization and Diversity in Dynamic Pattern Learning and Generation by the Distributed Representation Architecture (BDC-TR-2003-3)
Masato Ito, and Jun Tani
[ last update: 2003/11/11 ]
Learning to Generate Combinatorial Action Sequences Utilizing the Initial Sensitivity of Deterministic Dynamical Systems (BDC-TR-2003-4)
Ryunosuke Nishimoto and Jun Tani
- Abstract
This study shows how sensory-action sequences of imitating finite state machines (FSMs) can be learned by utilizing the deterministic dynamics of recurrent neural networks (RNNs). Our experiments indicated that each possible combinatorial sequence can be recalled by specifying its respective initial state value and also that fractal structures appear in this initial state mapping after the learning converges. We also observed that the sequences of mimicking FSMs are encoded utilizing the transient regions rather than the invariant sets of the evolved dynamical systems of the RNNs.
- Downloadable Materials
[ last update: 2003/11/19 ]
On-line imitative interaction with a humanoid robot using a dynamic neural network model of a mirror system (BDC-TR-2003-5)
Masato Ito and Jun Tani
[ last update: 2003/12/03 ]
[[ Misc ]]
This page is maintained by admin@bdc.brain.riken.go.jp
Book Chapter